Engagement Mechanics- Problems and Solutions

[Sabotin]

An Idea I've been juggling around my head:

 

Make melee equipped characters/monsters emanate an aura with a heavy slow effect instead of the soft-lock thing that there is now. The area depends on the range of the melee weapon and simulates this threat area existing now. Only affects enemies (relative to the melee equipped person). The slow effect doesn't stack with engagement areas from other characters (i.e. no double strength slow if you have 2 enemies engaging you)

 

Once you're caught you can walk away slowly (obviously only works if you're not being followed by the other character). If you have a skill that disengages you, you can use that and be moved outisde that particular engagement area or made immune to the slow. Every character would also have an options to force disengage (like the attack/cancel/talk to/guard buttons of IE games), which would function similarly as the aformentioned skills either moving you or making you ignore the slow for a second or two, but provoking a disengagement attack.

 

The attack is not provoked if there's multiple characters inside the same engagement area unless the engager has appropriate feats/abilities. The player can't be sure how many  caharacters need to be in the engagement area to not provoke a disengagement attack when force disengaging.

 

When you are inside multiple engagement areas you provoke a disengagement attack from all of them.

 

Reach weapons generate a larger engagement area, meaning you can engage people from further away, but also meaning that when fighting a non-reach weapon you can hit them before they get close enough to hit you (due to them being slowed from further away).

[Sensuki]

You're referring to something like a "melee shuffle". In and of itself it's not a bad concept, but it would definitely not constitute an Infinity Engine feel, and thus I will not support it.

[ISC]

An Idea I've been juggling around my head:

 

Make melee equipped characters/monsters emanate an aura with a heavy slow effect instead of the soft-lock thing that there is now. The area depends on the range of the melee weapon and simulates this threat area existing now. Only affects enemies (relative to the melee equipped person). The slow effect doesn't stack with engagement areas from other characters (i.e. no double strength slow if you have 2 enemies engaging you)

 

Once you're caught you can walk away slowly (obviously only works if you're not being followed by the other character). If you have a skill that disengages you, you can use that and be moved outisde that particular engagement area or made immune to the slow. Every character would also have an options to force disengage (like the attack/cancel/talk to/guard buttons of IE games), which would function similarly as the aformentioned skills either moving you or making you ignore the slow for a second or two, but provoking a disengagement attack.

 

The attack is not provoked if there's multiple characters inside the same engagement area unless the engager has appropriate feats/abilities. The player can't be sure how many  caharacters need to be in the engagement area to not provoke a disengagement attack when force disengaging.

 

When you are inside multiple engagement areas you provoke a disengagement attack from all of them.

 

Reach weapons generate a larger engagement area, meaning you can engage people from further away, but also meaning that when fighting a non-reach weapon you can hit them before they get close enough to hit you (due to them being slowed from further away).

 

This is actually a pretty good idea. I'd skip the whole force-disengage for everyone, and the disengagement attack, however. I think that adds unnecessary complexity, concepts and artificial mechanisms. I mean, if you have absolutely no talent, class or otherwise, for disengaging a swordsman, you will not be able to escape him easily. And by not having a disengagement attack the swordsman would not get two free (artificial) attacks in the middle of a normal attack when two targets try to move away from him. Instead he would have to chose if he would like to pursuit target A or B for "free" hits, or pursuit A while B is still hitting him etc.

 

Its kind of imperfect, but its a simple way to give fighters some control of the battlefield, to be able to block corridors, scare squishy rogues from coming to close etc, while maintaining the usefulness of disengage talents.

[Answermancer]

An Idea I've been juggling around my head:

 

Make melee equipped characters/monsters emanate an aura with a heavy slow effect instead of the soft-lock thing that there is now. The area depends on the range of the melee weapon and simulates this threat area existing now. Only affects enemies (relative to the melee equipped person). The slow effect doesn't stack with engagement areas from other characters (i.e. no double strength slow if you have 2 enemies engaging you)

 

Once you're caught you can walk away slowly (obviously only works if you're not being followed by the other character). If you have a skill that disengages you, you can use that and be moved outisde that particular engagement area or made immune to the slow. Every character would also have an options to force disengage (like the attack/cancel/talk to/guard buttons of IE games), which would function similarly as the aformentioned skills either moving you or making you ignore the slow for a second or two, but provoking a disengagement attack.

 

The attack is not provoked if there's multiple characters inside the same engagement area unless the engager has appropriate feats/abilities. The player can't be sure how many  caharacters need to be in the engagement area to not provoke a disengagement attack when force disengaging.

 

When you are inside multiple engagement areas you provoke a disengagement attack from all of them.

 

Reach weapons generate a larger engagement area, meaning you can engage people from further away, but also meaning that when fighting a non-reach weapon you can hit them before they get close enough to hit you (due to them being slowed from further away).

 

This is similar to what I've been thinking while lurkishly reading all the drama about engagement. but I think you could get rid of the attack entirely and just slow the movement of engaged enemies. Personally I like engagement as a goal, and I was happy to see the idea in the game, but I do agree that there are issues with it presently.

 

Ultimately I've been thinking that it should work more like zones of control in turn/tile-based games, meaning it stops/slows movement without the free invisible attack aspect. I honestly don't think we need free engagement attacks, I think that "tank" characters should emanate an area that seriously slows the movement of engaged enemies unless they use a "disengagement" ability. The area (and magnitude of the slow) could be affected by weapon reach and talents.

 

At that point the tank can prioritize who to actually attack if they try to get out of the engagement, and can also reliably land CC abilities like knockdowns and the like.

 

I'm not a huge fan of removing engagement altogether and relying entirely on CC abilities (as Sensuki advocates) but mostly just because of two reasons:

1. With recovery times, I think it requires too much micromanagement and luck to get a CC ability off at just the right time. In other words if there are enemies rushing past my "tank" I don't want to have to micromanage exactly what he's doing (making sure he's not in recovery at that moment) to get a CC ability off at just the right time. That just seems tedious to me, especially if there are multiple enemies. A slowing aura makes this much more manageable.

2. I'm worried that "boss"-type monsters will be able to ignore/shrug off CC and I don't think they should be immune to "engagement" of some sort unless that's a specific thing about that boss.

[Sensuki]

Ultimately I've been thinking that it should work more like zones of control in turn/tile-based games

This isn't a turn-based game. That is the entire problem with the mechanic. It's a product of turn-based design in the first place.

 

I'm not a huge fan of removing engagement altogether and relying entirely on CC abilities (as Sensuki advocates) but mostly just because of two reasons:

1. With recovery times, I think it requires too much micromanagement and luck to get a CC ability off at just the right time. In other words if there are enemies rushing past my "tank" I don't want to have to micromanage exactly what he's doing (making sure he's not in recovery at that moment) to get a CC ability off at just the right time. That just seems tedious to me, especially if there are multiple enemies. A slowing aura makes this much more manageable.

I don't know why this is so hard to understand, but as a player, using CC to control enemies will not be necessary if the AI targeting clauses were adjusted so that melee enemies turned to attack your units when you attack them in melee or something like that. If your problem is that you struggle to manipulate enemy targeting, a combination of positioning and understanding AI targeting will solve this problem completely.

 

 

2. I'm worried that "boss"-type monsters will be able to ignore/shrug off CC and I don't think they should be immune to "engagement" of some sort unless that's a specific thing about that boss.

There are no absolute immunities in PE, you will be able to Hobble a dragon.

Edited November 28, 2014 by Sensuki

[Osvir]

Problem:
- Space-pressing is way more frequent than in the IE games, a bit chaotic in a way, almost as if I press it too much in short intervalls. I think this has to do with combat animation speed across the board. And I feel animation should be slowed down a bit. I do like the Engagement mechanic though

 

Solution:
- Lower the attack speed animation. If the character starts swinging at 90% recovery in the BB, maybe they should start swinging at 70%-75%, and complete the animation at 100%. This might need a lot of testing and experimenting, but I believe should be fairly simple to manipulate, but might not be a priority at this development stage (I presume, bugs are more important to weed out).

A potential ally of this problem is the Combat Log, which is non-functional at this stage (at least for me) so I can only rely on the very well explained GUI. It suffices, but with a proper Combat Log that gives information at a fairly calm pace, I think it'd be easier to understand "X Player Attacks Y Target". "Aloth begins to cast Fireball" etc. etc. and it'd give a better idea what's happening too.

Right now the Combat Log is frustrating, it auto-scrolls to the very top, so I need to scroll it down every time, so I can't get a sense of the pace it's going at, but it looks like it's showing millions of numbers (exaggerated ofc) in a short spurt of time.

I suppose if it was simplified, with more basic numbers and basic actions, it could be easier to follow what's going on as well.

Example #1:

 

"X Player Attacks Y Target"

 

And then you could expand that to see that specific action in detail, being played out in realtime. The combat log could even stretch as far as have your characters names, with sub-groups that refreshes each combat. So if you want to know what one specific character is doing right now and have been doing up til this time, you'd expand that information by clicking on the name, and you'd see list you could scroll through whilst paused.

Example #2:

"Aloth"
- [Time]: Aloth attacks Dude
- [Time]: Dude took X damage
- [Time]: Dude attacks Aloth
- [CurrentTime]: Aloth took Y damage

"Edér"
- [Time]: Edér attacks Dude
- [CurrentTime]: Edér grazed Dude for Z damage

*shrug*

Problem #2:
- Characters are running everywhere, but when entering combat, they show that they have a walk animation at some times when it gets wonky. This affects the pace of the transition between out of combat and in combat a lot. You can sneak, and you can run, but you can't walk. In the IE games you were walking everywhere, in PoE you run everywhere. It affects the pace, and thus affects how you engage.

You literally run into battle when you Engage, and anywhere else for that matter (except when in cutscene).

Solution:
- Use the walk animation out of combat (IE feel), which already exists but only used in cutscenes, and either transist into running when entering combat, going for an enemy far away, or when trying to "run away" from Engagement (jRPG feel "Run Away"). Or Boots of Movement that activates "running in battle" but not out of battle (to not mess with pathfinding, and/or someone reaching the destination before everyone else). Or outright have a run-button that has some sort of penalty (Fatigue faster? Get "No-Rest-In-A-While" penalties faster? Engage battle with less Endurance? Etc.)

I think Pillars of Eternity would benefit a lot from having the walk animation instead of the run animation to use when traveling, personally, both out of combat, in combat and the transition between out of combat and in-combat would be better.

Try it Obsidian, in some build or whatnot, with a "Run or Walk" switch somewhere (console command/options) and see how the players react and which is most preferable, as a study of what is most played with and what the players enjoys the most, I suppose. What is immersion? What is convenience?

[Captain Shrek]

I would also like to see roll modifiers.

[Answermancer]

 

Ultimately I've been thinking that it should work more like zones of control in turn/tile-based games

This isn't a turn-based game. That is the entire problem with the mechanic. It's a product of turn-based design in the first place.

 

I get that, and I also understand that this is your problem with the mechanic as a whole. But I haven't seen anything that convinces me that the majority of people want to play this game the same way as an RTS. I'm sure there's some who do, but I'm not one of them and so have no problem with mechanics that stray from "this would work in an RTS game" being added when they make things I find frustrating easier to manage.

 

I mean, I'm not saying that my opinion is any more representative but I played the IE games with copious auto-pause and manual pausing, pretty much on every single fight, and I'm sure plenty of other people played them that way, so there's an immediate split there for what some backers want vs. others.

 

 

I'm not a huge fan of removing engagement altogether and relying entirely on CC abilities (as Sensuki advocates) but mostly just because of two reasons:

1. With recovery times, I think it requires too much micromanagement and luck to get a CC ability off at just the right time. In other words if there are enemies rushing past my "tank" I don't want to have to micromanage exactly what he's doing (making sure he's not in recovery at that moment) to get a CC ability off at just the right time. That just seems tedious to me, especially if there are multiple enemies. A slowing aura makes this much more manageable.

I don't know why this is so hard to understand, but as a player, using CC to control enemies will not be necessary if the AI targeting clauses were adjusted so that melee enemies turned to attack your units when you attack them in melee or something like that. If your problem is that you struggle to manipulate enemy targeting, a combination of positioning and understanding AI targeting will solve this problem completely.

 

I'm not following what you mean here about adjusting enemy targeting AI being a fix, can you elaborate?

 

And yes, to some extent my problem is that trying to perfectly time CC so that it goes off as enemies are trying to push past my front line feels very tedious right now (since you can't depend on such an attack going off as soon as you queue it up), and without some form of engagement I feel like too much of my time would be spent kiting enemies or micromanaging tank "recovery", which I don't find particularly fun.

 

I'd like to have more time to think about what abilities to use and where to move my people other than the front line, and having that front line provide something to facilitate that without a bunch of micromanagement.

 

This is why I like the idea of engagement being a passive slow. It provides me with a passive benefit to the front line that I can choose to micromanage further (or not) and something that the AI could likely be made to deal with more easily (ie. the choice of sticking to tank or moving through the slow, together with abilities that boost speed or counteract engagement).

 

 

2. I'm worried that "boss"-type monsters will be able to ignore/shrug off CC and I don't think they should be immune to "engagement" of some sort unless that's a specific thing about that boss.

There are no absolute immunities in PE, you will be able to Hobble a dragon.

 

You're right of course, I'm not sure why I forgot that.

[Jon of the Wired]

I don't know why this is so hard to understand, but as a player, using CC to control enemies will not be necessary if the AI targeting clauses were adjusted so that melee enemies turned to attack your units when you attack them in melee or something like that. If your problem is that you struggle to manipulate enemy targeting, a combination of positioning and understanding AI targeting will solve this problem completely.

 

I don't see this as a solution because I want the enemies to be smart, unpredictable, and generally behave more believably like people (or monsters, or what-have-you). The current AI doesn't achieve that goal, but if you design the game around stupid, manipulable AI, then you never have the option to make it better.

[Sensuki]

I'm not following what you mean here about adjusting enemy targeting AI being a fix, can you elaborate?

Sure.

 

Melee Engagement consists of two components - an AI targeting clause, and a Disengagement attack.

 

When a unit (player or AI) engages another unit of the opposing faction (AI or player), the unit that is engaged stops, turns and attacks their engager. This is done through an AI targeting clause. If that AI targeting clause didn't exist, then the engaged unit wouldn't stop to attack.

 

This is the key part of the Melee Engagement system for 99.99% of people that enjoy it, including Shevek and Kjaamor. People like that it gives them easy control over the enemy AI.

 

The thing that I find annoying about this is that it overrides the actions of my units, and I am not a fan of anything that overrides my unit actions.

 

The Melee Engagement system does not have to exist for this behavior to occur. The Enemy AI can be programmed to stop and attack the first melee unit that they are attacked by.

 

I don't see this as a solution because I want the enemies to be smart, unpredictable, and generally behave more believably like people (or monsters, or what-have-you). The current AI doesn't achieve that goal, but if you design the game around stupid, manipulable AI, then you never have the option to make it better.

I want the AI to be smart too, however it cannot be with the Melee Engagement system or with people who want to easily be able to manipulate what the enemy AI are doing. I agree with you, I want to put in effort to control the battlefield, with unit positioning and movement - like in IWD:HoW or even better, which had the best targeting of any of the IE games.

 

That is not what we will get though. Obsidian has decided to cater to the people who find it a chore to control units in combat, and created a system so that they can easily manipulate the AI with some basic initial movements, at the cost of tactical movement and positioning in combat after the opening.

 

Melee Engagement overrides all other targeting clauses, and it makes no sense for enemies to randomly disengage because they suffer disengagement attacks, so once enemies are engaged by you first, they will attack you until they or you die.

Edited November 28, 2014 by Sensuki

[Shevek]

Or you can have engagement and give certain skilled enemies (in set piece encounters or in a few tough trash encounters) the same engagement breaking abilities/tactics as the player (knockdowns, stuns, etc). That would give encounter design an added dimension and help differentiate encounters/enemies. You can have dynamic set piece encounters and casual friendly trash encounters. Engagement has the possibility of adding quite a bit to the palatte OE can paint with when crafting this game.

 

Or we can throw the baby out with bathwater and go back to funneling all enemies through doors.

[Sensuki]

Not really. Single target abilities and spells that can break engagement only have a chance to hit and they don't work against multiple engagements. AoE abilities that break engagement suffer from the same problem and they're quite rare. All the player has to do to circumvent these is to learn when they occur and circumvent them. What's more, the amount of programming required to get AI to work as a team like that would be far beyond the scope of the project - stuff like that is even a problem for Valve with the DotA 2 bots.

 

You're thinking about it from the wrong angle, as always.

 

edit: Forgot to add, it doesn't really make sense for the AI to disengage from the player in either situation. The AI does not care whether they live or die, the purpose of enemies that don't have cutscene scripted escapes or whatever are to inflict as much hurt on the player as possible. Once they are in melee combat with an enemy, they should stay there unless that enemy dies, or they lose their target (via whatever means).

Edited November 28, 2014 by Sensuki

[Shevek]

Lol, you claim to want great AI and then you so say no great AI is not possible. Blah blah blah. Your tripping over yourself, man.

[Sensuki]

Nope, you made a non-sensible request.

 

As I said, melee AI has no need to disengage from the player.

 

Smart, deterministic AI would be to have intelligent AI targeting clauses, with good scripted ability use and good target reacquisition. Melee Engagement removes most of the need to have robust targeting.

 

Why bother when most of the time the player will override it with engagement?

Edited November 28, 2014 by Sensuki

[Shevek]

Uh huh, I dont hear an argument. I hear nonsense.

[Sensuki]

Perhaps it's simply beyond your comprehension ?

[Shevek]

Ya, maybe its like Kubrik or Lynch movie. I makes sense if you are sufficiently loaded.

[Answermancer]

 

I'm not following what you mean here about adjusting enemy targeting AI being a fix, can you elaborate?

Sure.

 

Melee Engagement consists of two components - an AI targeting clause, and a Disengagement attack.

 

When a unit (player or AI) engages another unit of the opposing faction (AI or player), the unit that is engaged stops, turns and attacks their engager. This is done through an AI targeting clause. If that AI targeting clause didn't exist, then the engaged unit wouldn't stop to attack.

 

This is the key part of the Melee Engagement system for 99.99% of people that enjoy it, including Shevek and Kjaamor. People like that it gives them easy control over the enemy AI.

 

The thing that I find annoying about this is that it overrides the actions of my units, and I am not a fan of anything that overrides my unit actions.

 

The Melee Engagement system does not have to exist for this behavior to occur. The Enemy AI can be programmed to stop and attack the first melee unit that they are attacked by.

 

I see, that sounds okay the way you describe it, though I wonder how it would feel in practice. Even if enemy AI attacked the first melee unit they were attacked by, wouldn't they potentially just run off right away if someone else hit them (a squishy ranged charater for instance)? If so it doesn't really give you more time to land CC (what I would like at minimum from front liners), unless the script had some kind of "hit this guy at least once before switching targets" clause.

 

Assuming such a clause did exist, wouldn't this still go against the way you want to play, even if it did keep your own units from having their actions overriden (which you've identified as your biggest issue with engagement, I think)?

[Sensuki]

No, it would be first attacked by. Currently it's first engaged by. If an enemy is engaged by multiple characters, they attack the first one. In the case of first attacked by, if they were attacked by multiple enemies it would be first melee enemy they were attacked by.

 

I'll make a video this week and show you how it looks without Engagement (I have modded it out).

Edited November 28, 2014 by Sensuki

[constantine]

How about this topic gets a poll-

 

 

Do you want the engagement mechanic in Pillars of Eternity ?

 

Yes

 

No

[ISC]

Only if the poll comes with key arguments for and against. I dont think everyone realizes what Sensuki just explained.

[Sensuki]

I don't create polls. Opinions and reasons for those opinions are more informative than a simple yes or no.

[Captain Shrek]

I have a feeling that way too much time is being spent discussing the engagement mechanics. At some point in time you have to ask  if that is so much of an issue. I believe that removing it or keeping it, does not really change so much, It's not going to rescue PoE combat from its mind-numbing mediocrity. 

 

I would seriously reconsider other aspects of the game, like combat speed and lack of out of combat magic, buffing etc over this triviality. 

Edited November 30, 2014 by Captain Shrek

[Sensuki]

Having done some testing in the new build, it appears that attack interaction with moving targets has been improved at least for certain weapons, and there is now target re-acquisition and it plays very very well removed from the game. Need to do more testing, but it's looking very promising.

[constantine]

But for maybe the dozenth time, I will say my opinion for engagement:

 

I like the thought of it and the fact that it's there, but the execution is wrong beyond measure.

 

 

Please tweak it.

 

Suggestion: have disengagement attacks happen on a cooldown, not indefinitely.

 

It is a very serious issue that movement-maneuvering in combat is so hampered.

 

Have the bold (or stubborn) take the hit and charge on, not be massacred right there in the spot.

 

Why is this so much to ask ?

Edited November 30, 2014 by constantine